Ask people what they want and they will sketch you a faster horse. Research craft is not interrogation — it is piecing back a concrete past episode, minute by minute, until the genuine constraints appear.
Ask about the last time, not the next
“Would you use this?” invites invention. “Talk me through the last time you attempted this” invites recall — messier, yet honest. Tie each question to one recent, specific occasion and trace its timeline.
Five habits that shift everything
First: hold silence after an answer — count to five before you speak. Second: ask which screen sat in front of them. Third: ask who else played a part. Fourth: ask what they tried just before quitting. Fifth: close with “what should I have asked?” — people regularly hand you the study you ought to have run.
Frame findings as wagers, not verdicts
Five conversations prove nothing; they lower the risk of the next call. Phrase each insight as “we suspect X because we observed Y; we will know we are right when Z” — and research feeds the roadmap instead of yellowing inside a slide deck.
